INDEX

Explanations

adjectives or adverbs indicating a high degree or intensity

oai_token-act-pair · gpt-3.5-turbo

the word "so" indicating emphasis or intensity

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wreck

-0.60

 glances

-0.60

 eviction

-0.59

 acre

-0.57

ropolitan

-0.55

 gallery

-0.54

theless

-0.54

{:

-0.53

haven

-0.53

idon

-0.53

POSITIVE LOGITS

bered

1.19

oths

1.12

othes

1.11

oooo

1.05

apy

1.04

ooo

1.02

othe

1.01

oner

0.99

zin

0.99

aps

0.96

Activations Density 0.126%

adjectives or adverbs indicating a high degree or intensity

the word "so" indicating emphasis or intensity

No Comments

No Known Activations